Shapecaret-double-leftcaret-double-rightcaret-downcaret-leftcaret-right-circlecaret-rightShapeclosedropdownexpand morefacebookLogolinkedinlogo-footerlogo-marklogo-mobilemailsearchtwitteryoutube
Menu Close

Clinical Data Standardization

Developing and bringing a new product to the market in personal care and foods very often includes a clinical trial for assessing the efficacy of the product and supporting claims. Using effective technology-enabled methods to manage clinical data can enhance the speed with which the product is developed and commercialized, hence enhancing the competitive advantage. Data can come from various sources like subjects characteristics, answers to specific questions, measurements with different instruments, amount of product usage, etc. "How can we standardize the data collection and store the data?" was the question Velitchka was asked when she started her project at Unilever.

One can compare data standardization to building LEGO pieces: most of them have a regular shape and can be used to build different things. We first looked at the most commonly used measurement and assessment methods for which we defined what data needs to be gathered and how it will be stored. Soon we had a large collection of standards (LEGO pieces) we could reuse depending on the particular clinical trial design. With statisticians we defined dataset that were used for the statistical analysis. We went a step further and defined an automated procedure that will process all the data and make them available for analysis.

Is there more to it?

Standardized data are of high quality, consistent and can be retraced to the original raw data. But "Is there more to it?" one may ask. And the answer is "Yes". It is now possible to combine data from multiple studies and gain insights without the need of conducting a clinical trial. When information on product composition is available one can build models for predicting product efficiency and use in the clinical trials only the most promising products. Current developments in computational algorithms together with high quality standardized data are fundamental for creating in silico tools that make possible to say whether a product will work and will make financial sense before it even exists in the real world.

What's your next step? We can help you with that